| thumbnail: "https://deej-ai.online/favicon-96x96.png" | |
| tags: | |
| - audio | |
| - music | |
| license: "gpl-3.0" | |
| This model encodes audio files into vectors of 100 dimensions. It was trained on a million Spotify playlists and tracks. The details can be found [here](https://github.com/teticio/Deej-AI). | |
| To encode an audio first install the package with | |
| ``` | |
| pip install audiodiffusion | |
| ``` | |
| and then run | |
| ```python | |
| from audiodiffusion.audio_encoder import AudioEncoder | |
| audio_encoder = AudioEncoder.from_pretrained("teticio/audio-encoder") | |
| audio_encoder.encode(<list of audio files>) | |
| ``` | |